Compress PDF

How to Compress PDF Files?

Before we start talking about compressing PDF files, we have to understand what a PDF file is and how it’s constructed. Think of a PDF file as a container that contains different elements: text, images, and structure (these are fonts, metadata, bookmarks, fields, hidden data from the PDF editor).

What does “Compressing” a PDF actually mean?

So compressing a PDF entails shrinking that container without changing what the reader sees, and there are three levels to achieve optimal compression.

1. Shrink the images. Photos are always the biggest chunk of a PDF’s size. Re-saving them at a lower quality or smaller resolution is the biggest lever we can pull to shrink the file size.
2. Strip what nobody reads. These include the author name, editing history, thumbnail, fonts that aren’t used, or hidden attachments. Deleting this won’t affect what the user sees.
3. Rewriting the file cleanly. A PDF file collects junk during repeated edits. So rebuilding the internal structure removes any gaps or duplicates.

That’s the simplest explanation of how PDF compression works, and nearly every tool on the internet pulls one of these levers.

What to look for before compressing a PDF file?

Before compressing a PDF, here are the things to consider.

Do you need the text to be searchable or selectable? For most users, the answer would be yes. So keep it as text.

Do you need it compressed to a specific size? File attachments using email usually cap at 25 MB. Government portals often cap between 2 and 5 MB. If this is the case, choose a target size; don’t guess.

How is it viewed? On smartphones, PDF files at 150 DPI look similar to 300 DPI. However, if you print it, the difference will be noticeable.

If the file you’ll send is only read on screen and emailed, you can choose a lower resolution (150 DPI is a good compromise). If it’s going to be printed out, it should be at least 300 DPI.

How does ConvertYard compress your PDF?

ConvertYard compresses PDF files in the browser. This means that nothing is uploaded to a server; nothing leaves your computer. When you open or drop a PDF file, here’s what happens:

1. WebAssembly processes the PDF file within the browser.
2. It rewrites the internal structure to remove any excess space (that’s equivalent to bytes).
3. Metadata is stripped by default (this means the author name, title, and editing history are deleted).
4. If you choose this option, it re-encodes the images at a lower quality.
5. Rewrite the new PDF file, ready for download.

If by any chance the compressed version has a larger file size than the original (this happens if the original file uses compressed images or is optimized), we return the original file instead. You don’t get a worse PDF file than what you have.

What to expect from this compressor?

1. Scanned documents or image-heavy PDFs: expect around 40 to 80% compression. This is where this compressor excels.
2. Exported reports with charts and photos: expect around 20 to 50% reduction.
3. Text-only PDF: expect around 0 to 15%. There’s not much to compress here.
4. Already-compressed PDFs: expect little or no change. If it’s already compressed, there’s nothing left to squeeze unless you want the images to be heavily pixelized.

Compression will take a few seconds on smaller files, and a few minutes with larger files. You can compress PDF batches of up to 100.

Basic Features

Compression level: Low, Medium, High, Aggressive. Low is the safest option but offers the least compression. Medium is the default option. High will re-encode images at 30% quality. Aggressive flattens the PDF file and turns every page into an image, so text will not be searchable. Best used for scanned PDF files.

Target size mode: KB or MB with presets between 100 KB and 10 MB. For example, if you choose a 2 MB compression, the page runs multiple compression cycles and automatically stops when it reaches the target. This works well on image-heavy PDFs, but not as well on text-heavy PDFs.

Batch compression: Users can compress between 1 and 1,000 files per batch. PDFs can be downloaded individually or as a batch (ZIP file).

Advanced Features: If You Need More Control

Custom DPI: Force downgrade to a specific resolution. 72 DPI for email attachments. 150 DPI for viewing on screen only. 300 DPI for print quality.

JPG Quality Slider: Choose between 10 and 95 for embedded images. 70 is the sweet spot to balance compression without losing quality.

Grayscale: Turns a colored PDF into black/white (or grayscale). Cuts image size by around 60%.

Strip options: Remove metadata, annotations, bookmarks, embedded files, or JavaScript.

Remove unused fonts: Removes any font that isn’t used by the PDF file. Doesn’t affect readability of the file.

Flatten or remove form fields: Flattening or removing the form fields will not affect the way the PDF file looks, but the file can’t be edited anymore. This option will remove the fields.

Strip private app data: Removes any data left behind by Photoshop, Illustrator, Word, and other tools.

Why are file uploads risky for PDFs?

Here are some examples: Tax returns that have your Social Security Number. Medical records. Signed contracts with your signature. A passport with your photo, birthday, and passport number — information hackers can use to steal your identity. Signed contracts with your signature. Scanned documents you sent to your landlord.

When you upload these types of files into a “free online PDF compressor,” they are dropped into a server owned by a company you don’t know. These websites don’t disclose where they store uploaded PDFs. Some give you that option to delete the PDF files, but most users won’t bother.

Who can read it? A number of people, including server admins, third-party analytics tools, basically, anyone with enough credentials to open these files. The marketing copy on the website says “secure,” but that’s only a claim, not a guarantee.

How long does it stay uploaded? That would depend on the website. Some say files are deleted after an hour. Others? Thirty minutes. It varies. However, one thing they don’t disclose is whether they automatically back it up for redundancy. This may also include these PDF files. Again, these aren’t disclosed.

Where does it end up? Data is usually stored on multiple servers distributed across different data centers. This is to serve data at the shortest possible distance for efficiency. This redundancy helps with the website’s core files, but it doesn’t protect PDF files: if a misconfigured cloud bucket exposes your data, hackers can access it. Try searching for “PDF converter data leak” to see what I mean.

Who’s training on it? Most people don’t realize that there may be some free tools that are quietly passing your uploads directly into AI training pipelines. Your PDF contract may be in a model’s next training set.

PDFs are worse than images because of the type of data that could leak. A photo of your logo for a client could be copied by another graphic artist. The worst-case scenario is that you create another design. A leaked tax return can lead to identity theft. A leaked NDA is a lawsuit waiting to happen. The stakes are higher and costlier. Every upload is a risk that could cost you thousands of dollars.

ConvertYard sidesteps all this because it doesn’t upload anything. The PDF opens, compresses, and saves inside the browser. Once you close the tab, the file disappears. Want proof? Open Google Chrome, select ‘more tools’ > ‘developer tools’, and choose the network tab. Run a compression and watch for a request that carries a file. You won’t see any. Do the same with a cloud-based compressor.

So if a document is sensitive enough that you’d need to shred the hard copy, then it’s sensitive enough that you shouldn’t upload the PDF version.

How ConvertYard compares